Mining Operations Turn to AI to Manage Supply Chain Risks
Researchers at the National Laboratory of the Rockies are building artificial intelligence models to optimize how mines extract and process critical minerals - materials essential to everything from smartphones to aircraft.
Ryan King, a computational scientist at the lab, is working to inject flexibility into mining operations that have remained largely unchanged for decades. Heavy equipment installed in mines often stays in place for generations, but the minerals those operations need to extract shift with geopolitical events, technological advances, and resource depletion.
"Supply chains evolve rapidly, and we need processing systems that can absorb input volatility," King said.
Real-World Data Accelerates Model Development
King's location in Minnesota - the largest U.S. iron ore producer - gives him direct access to mining operations and expertise. Through a partnership with the University of Minnesota's Natural Resources Research Institute, his team is training AI models on actual operational data from iron ore processing.
The work focuses on adjusting processing steps based on the intended end product. Different applications require different ore purities, and AI can optimize those decisions faster than human operators working with fixed procedures.
King came to this work through a background in fluid dynamics and mechanical engineering. He began applying AI to scientific problems during his doctoral studies while interning at the lab in 2012.
AI Across the Supply Chain
The National Laboratory of the Rockies is contributing to the U.S. Department of Energy's Genesis Mission, a national initiative to strengthen critical minerals supply chains. King leads the AI seed model team for the Critical Minerals and Materials To Unlock Supply (CM2US) program.
The goal is to integrate AI into every step of the mining process - from exploration and extraction through milling, separation, and reduction. AI models can identify connections across these stages that humans cannot optimize at the required speed.
This approach offers two operational advantages: predicting supply disruptions before they occur, and rapidly testing material compositions to identify the most promising options for lab work.
"AI can plug in anywhere in the process," King said. "It can help with exploration and extraction, and then optimize the design and controls of key processing steps."
Scale and Access
Genesis Mission will integrate expertise from 17 national laboratories into a single platform. This gives researchers across the country access to capabilities that can improve mining operations beyond Minnesota's Iron Range.
The methods King develops for iron ore processing apply broadly to other minerals. The AI challenges in mining - managing variable inputs, optimizing outputs, and responding to changing market demands - remain consistent regardless of what mineral is being extracted.
Learn more about AI for Operations and AI for Science & Research.
Your membership also unlocks:
